Yuchen Zhu
Machine Learning PhD @ Georgia Tech 🍀

140 Skiles
686 Cherry St NW
Atlanta, GA 30332
Hi, I am Yuchen Zhu, a 2nd year Machine Learning PhD at Georgia Tech.
My interests lie in the broad aspects of GenAI, wich applications in vision, language, and sciences (e.g. single cell genomics, protein sciences, etc). My research currently focuses on diffusion models/flow-based methods (especially discrete diffusion), and multimodal foundation models (such as VLMs, Multimodal LLMs).
At Georgia Tech, I am fortunate to be advised by Molei Tao and Yongxin Chen, and working with a group of incredible researchers. Prior to that, I graduated with MA in Statistics from Yale University and BS in Honors Mathematics with highest honor from NYU Shanghai. During those time, I had the privilege to work with Zhuoran Yang and Mathieu Laurière on theory/numerics of RL and mean-field system.
Contact: yzhu738 [at] gatech [dot] edu
Updates
[06/2025] Mimicking or Reasoning is public! Check out our new work on evaluating MM-ICL for VLM reasoners! |
[06/2025] Diffuse Everything is online! Check out our new work on multimodal diffusion models on native state spaces! |
[05/2025] Learning to Stop and Diffuse Everything got accepted to ICML 2025, see you in Vancouver! |
[04/2025] I wrote a new blog on how group structures aid generative modeling of manifold data. |
[02/2025] Check out my new work on fast high-order samplers for discrete diffusion models! |
[01/2025] TDM and STEM got accepted to ICLR 2025, see you in Singapore! |
Talks
[11/2024] GT ML Student Seminar |
[10/2024] SIAM Mathematics of Data Science 2024 Atlanta, talk and poster presentation |
[04/2024] Southeast ACM Student Workshop 2024, student presentation |
Selected Publications
- Mimicking or Reasoning: Rethinking Multi-Modal In-Context Learning in Vision-Language ModelsPreprint, 2025
- Diffuse Everything: Multimodal Diffusion Models on Arbitrary State SpacesInternational Conference on Machine Learning (ICML), 2025
- Learning to Stop: Deep Learning for Mean Field Optimal StoppingInternational Conference on Machine Learning (ICML), 2025